diff --git a/vllm/attention/ops/triton_flash_attention.py b/vllm/attention/ops/triton_flash_attention.py index e98b5254541b6..23ac7d7dc84ca 100644 --- a/vllm/attention/ops/triton_flash_attention.py +++ b/vllm/attention/ops/triton_flash_attention.py @@ -650,7 +650,7 @@ def get_general_autotune_configs(): def has_cdna_target(): - ROCM_CDNA_TARGETS = ["gfx940", "gfx941", "gfx942", "gfx90a", "gfx908"] + ROCM_CDNA_TARGETS = ["gfx942", "gfx90a", "gfx908"] return triton.runtime.driver.active.get_current_target( ).arch in ROCM_CDNA_TARGETS